Skip to main content.

Introduction

The assessment of the state of fresh waters in New Zealand and worldwide has been dominated by one-off chemical and field metered snapshots. Analysing flowing water using this approach has clear limitations as these measures vary seasonally, daily and even hourly and moreover they are all flow and rainfall influenced. This inherent variability means that long-term trends are missed due to the difficulty of finding statistically significant trends from the natural variability in such data. While these snapshots may be useful for later diagnosis they are not suitable for assessment of ecosystem condition. One solution to this problem is to use the biota (biomonitoring or bioassessment) because the resident life forms integrate all possible snapshot “water quality” measures over meaningful time scales. This realisation has changed the approach taken by many developed countries as epitomised by the European Unions Water framework directive1.

One aspect of the flowing freshwater biota particularly suited to biomonitoring is fish; they are long lived, relatively easily captured and are highly valued by the community. At the heart of bioassessment worldwide1 is the comparison between the biota that would be at a site in the absence of human impact with what is found there now. This can be achieved in a number of ways; either through predictive models that allow for site specific predictions and comparisons or less directly through an index that has multiple metrics and is calibrated to local conditions.

When making comparisons between sites the number of species can only be used if there has been repeat sampling at that reach over time. There has been a clear consensus that distribution of New Zealand freshwater fish fauna is driven largely by elevation and distance from the coast due to the prevalence of migratory fish species . This results in the number of species decreasing with increasing distance from the sea and elevation.

To enable between-site comparisons it is necessary to use a method that takes this into account. A Fish Index of Biotic Integrity (F-IBI) can achieve this requirement as fish community expectations are based on elevation, distance and existing conditions.

Using fish communities to define freshwater ecosystem integrity has had a relatively long history. The Index of Biotic integrity using fish was originally developed in the USA by James Karr during the early 1980s and is now used worldwide . A F-IBI has been developed for all New Zealand and regionally for the Waikato , Auckland , Hawke's Bay and Wellington . The index of biotic integrity allows for the assessment of the fish community at a site that is independent of the elevation or distance from the sea. These two parameters are critical as the majority of the fish fauna in New Zealand is migratory. The IBI score is based on the sum of 12 scores for different aspects of fish communities; the maximum value is 60 and the minimum is 0 if there are no native fish present.

In the last year the fish IBI has been further modified for the Auckland and Waikato Regional Councils with the inclusion of a recent statistical development; quantile regression (Chen 2005). This change resulted in more accurate scoring of sites but also meant that IBI scores were higher for many sites than previous IBI indices. This is because the previous process of fitting lines by eye overestimated the slopes of lines as the densities of data points could not be accurately ascertained. The new quantile IBI approach was used in the analysis for this report.

The aim of this report is to identify national trends in freshwater fish communities over the past four decades, as an indicator of ecosystem condition, and identify any relationships between trends in fish communities and land use.

 

1 In Europe, the European Water Framework Directive (WFD2000/60/EC; European Commission, 2000) introduces the obligation for its member states to achieve and maintain good ecological status for all water bodies. Such ecological status must be assessed as a deviation from the reference condition, measuring the Ecological Quality Ratio (EQR = Observed/Expected) for different quality elements (macroinvertebrates, diatoms, macrophytes and fish). While some European countries have adopted the multimetric approach in assessing biological communities (e.g. ), others have chosen to develop predictive systems based on the multivariate approach.

 

[ |